COMMUNICATIONS TO THE EDITOR
Dec. 20, 1958
6691
P, 10.00. Found: P, 9.78). Di-n-octylphosphine (V) secondary deuterium isotope effect ( k H / k D = oxide, m.p. 85-6' (lit.6m.p. 85') was characterized ca. 1.5) with formic-d acid, undoubtedly due to by its reaction with acrylonitrile to give 2-cyano- stretching of the carbon-hydrogen bond during ethyldi-n-octylphosphine oxide, m.p. 51-52' (lit.' the slow step, B, and (VI) carbon monoxide of Bis-(2-~yanoethyl)-phosphinenormal isotopic composition formed during reaction m.p. 53.4-54.2'). oxide, m.p. 98-99', (calcd. for CsHsNzOP: C, 46.15; in oxygen-18 enriched sulfuric acid. Step C fails H , 5.80; P, 19.84. Found: C, 46.32; H, 5.95; to occur because of its unfavorable competition with PI 19.94) was characterized by reaction with chloral D, the rapid proton transfer to the sulfuric acid. hydrate to give bis-(2-cyanoethyl)-l-hydroxy-2,2,2- For the triphenylacetic acid decarbonylation the trichloroethylphosphine oxide, m.p. 159-160' dec. scheme, A, B, C, D, E, adequately accounts for (calcd. for C6H1&13N202P: C, 31.65; H, 3.32. facts I and 11. Some back reaction of the oxoFound: C, 31.76; H, 3.35). Additional examples carbonium ion with water, C, can occur by favorand experimental details will be given in a sub- ably competing with D which is an attack by sulsequent publication. furic acid on the hindered number 2 carbon atom STAMFORD LABORATORIES M. M. RAUHUT with ejection of carbon monoxide, and which is understandably slower than the analogous proton RESEARCH DIVISION I. HECHEXBLEIKNER AMERICANCYAKAMID COMPASY HELENA . CURRIER transfer in the formic acid decarbonylation. Since V. P. WYSTRACHthe oxygen-18 study indicates that C is slower than STAMFORD, CONXECTICUT RECEIVED OCTOBER 25, 1958 D, however, the mechanism is closer to that of formic acid3 than to the other extreme proposed ISOTOPIC EVIDENCE FOR THE MECHANISMS OF by Den0 and Taft.l The reported5,6 non-integral DECARBONYLATION OF THREE CARBOXYLIC ACIDS slope of the plot of log k vs. HOmay be due t o the IN SULFURIC ACID intermediate character of the mechanism with Sir : neither B nor D strictly rate controlling. As reaction mechanisms are often determined The proposed mechanism6 of decarbonylation of from the form of the observed acid catalysis, it is benzoylformic acid can explain the results of isosignificant that the mechanism' of decarbonylation topic studies: (VII) the large isotope effect7 with of triphenylacetic acid proposed on this basis is not benzoylformic-l-C14 acid (k12/k14 = ca. 1.1) due to supported by isotopic evidence. Complete equi- carbon-14-oxygen bond cleavage in the rate step, librium of the oxocarbonium ion with water1 and VIII. A smaller effect (k12/k14 = cu. 1.036) with carbon-carbon bond cleavage as the rate determin- benzoylformic-2-CI4 acid, probably due to the effect ing step' are not supported since: (I) this decar- of isotopic substitution a t the number 2 carbon on bonylation in oxygen-18 enriched 95 .5YGsulfuric acid the equilibrium constant of the reversible protonayielded carbon monoxide having an oxygen-18 tion involving the alpha-keto group, and IX. enrichment of only about one-fifth that of the unenriched carbon monoxide from decarbonylation sulfuric acidI2 and (11) no measurable isotope of benzoylformic acid in oxygen-18 enriched effect was found in the decarbonylation of tri- sulfuric acid, reasonable by analogy with the phenylacetic-2-C14 acid. formic acid decarbonylation mechanism. For the decarbonylation of formic acid the Non-radioactive carbon monoxide from decarmechanism, bonylation of benzoylformic-2-C14acid confirmed an 0 earlier report* that the carbon monoxide came only A . RCOOH f H2S04 RC/ +- HS0@ from the carboxyl group. \OH - ~Helpful suggestions of John D. Roberts and F. A. H@ Long are acknowledged. slow B. RCNo RCOe f H20 \OH H@ C. RCOe f H?O + RCJ
0
\OH@ H D. RCOe f H2S04 + (R.HzS04)e f CO work2Hz0 -+ ROH f H30@ f H2SO4 E. (R.HzS04)e
+
UP
accounts for the facts: (111) log of pseudo-first order rate constant linearly related to Ho3; (IV) Large carbon-14 isotope effect (k12/k14 = cu. 1.09),4 (1) S . C. Deno and R. W. Taft, THISJOURNAL, 76,248 (1954).
(2) Although the two oxygen positions in carboxylic acids are not exactly equivalent, they should rapidly equilibrate through the symmetrical form, R-C'OH.
@\OH
Hence. the proposed' equilibrium involv-
ing the oxocarbonium ion and water should result in completely enriched carbon monoxide. (3) L. P. Hammett, "Physical Organic Chemistry," McGraw-Hill Book Co., New York, N. Y., 1940, p. 283. (4) G. A. Ropp, A. J. Weinberger and 0. K. Neville, T a r s JOURNAL, 73, 5573 (1951); H. Eyring and F. W. Cagle, Jr., J . Phys. Chem., 889 (1952).
(5) L. P. Hammett, Chem. Revs., 16, 67 (1935). (6) W. W. Elliott and D. L. Hammick, J. Chem. SOL.,3402 (1951). (7) B. Fingerman and M. Lemmon, Bio-Organic Chemistry Report, Radiation Lab., Univ. of California, Berkeley, Calif., 1958, UCRL-
8204. (8) K. Banholzer and H. Schmid, Hela. Chim. A d a , 39,548 (1956). (9) Chemistry Division, Oak Ridge National Laboratory, Operated by Union Carbide Corporation for the U. S Atomic Energy Commission.
OAK RIDGE?TENNESSEE
Gus -4.ROPP
RECEIVED XOVEMBER 3, 1958 INTRAMOLECULAR HYDROGEN BONDING INVOLVING DOUBLE BONDS, TRIPLE BONDS AND CYCLOPROPANE RINGS AS PROTON ACCEPTORS
Sir: We wish to report evidence which demonstrates the occurrence of intramolecular hydrogen bonding between proton donors and unsaturated linkages, including cyclopropane rings, as proton acceptors. Recently, similar observations have been reported for intramolecular interactions between hydroxyl